Urban forest biodiversity and cardiovascular disease: Potential health benefits from California’s street trees

work, commuting, etc.) the green characteristics of that landscape would be a more notable determinant of well-being [5]. Perhaps more subjective than any measure of psychological restoration would be a space’s stated aesthetic value, which plays a central role in the extent to which the space improves an individual’s subjective well-being [43]. Aesthetics, along with any subjective benefit derived by humans from biodiversity, does not follow a positive relationship with diversity measures. Aes- thetic value is not as much a pure function of diversity as it is a product of one’s schema of what makes a green space attractive and enjoyable. Qiu et al. [44] find that the extent of one’s ecological knowledge serves as another moderator that shapes one’s perception of, preference for, and enjoyment of biodiversity. The final mechanism by which biodiversity may affect human health is through the “dilu- tion effect” hypothesis [16,45]. This hypothesis states that high vertebrate biodiversity reduces the risk of humans contracting diseases as the greater species mix dilutes the concentration of vertebrates often carrying harmful diseases transmittable to human-beings through an arthro- pod vector (i.e. Lyme Disease) [9]. However, evidence for the dilution effect is mixed when meta-analyses restrict to field studies [46]. Taken together, diversity within a stand of trees could enhance human immune functioning beyond that of a stand of a single genus through the commensal human microbiome. It could

also improve restoration, reduce stress, and reduce exposure to arthropod-borne disease. However, it could also be insignificant in modifying the immune system, or detrimental to human health through some alternate pathway or reservoir [41]. Furthermore, since biodiver- sity is positively correlated with the number of trees in a forest, the separate effect of biodiver- sity from exposure to natural trees, as well as their interaction, is unclear. Our null hypothesis is that there is no significant relationship between biodiversity of urban trees, conditional on exposure to trees. The alternative hypothesis is that urban biodiversity is associated with human health outcomes. Materials & methods Study area This study includes zip codes in the U.S. state of California from 2010–2018. California has 1,010 zip codes. Zip codes included in the analysis are the 857 zip codes with data available for each year for our Heart Disease sample and the 241 zip codes with data available for each year for our Stroke sample. Data sources Our analysis uses data on health outcomes, urban forest characteristics, socio-demographic characteristics, and pollution characteristics. Table 1 presents summary statistics for the pri- mary variables used within our analysis. For summary statistics on additional control variables used in our sensitivity analysis, see S1 Table in Supporting Information. Table 1. Summary statistics. Mean SD Min Max N Heart Disease 169.6 84.74 26.28 1,357 7,713 Stroke 49.30 28.54 0 657.7 7,713 Shannon Index 2.741 0.643 0 3.830 7,713

Tree per 100K Pop 0.158 0.178 6.57e-05 2.196 7,713 https://doi.org/10.1371/journal.pone.0254973.t001 PLOS ONE Biodiversity and cardiovascular disease PLOS ONE | https://doi.org/10.1371/journal.pone.0254973 November 3, 2021 4 / 23